EP1712029B1 - Methods and apparatuses of DATA TRANSMISSION WITH SPATIAL SPREADING IN A MIMO COMMUNICATION SYSTEM - Google Patents

Description

BACKGROUND II. Field
• The present invention relates generally to communication, and more specifically to techniques for transmitting data in a multiple-input multiple-output (MIMO) communication system.
III. Background
• A MIMO system employs multiple (NT) transmit antennas at a transmitting entity and multiple (NR) receive antennas at a receiving entity for data transmission. A MIMO channel formed by the NT transmit antennas and NR receive antennas may be decomposed into Ns spatial channels, where Ns ≤ min {Nτ, NR}. The Ns spatial channels may be used to transmit data in parallel to achieve higher throughput and/or redundantly to achieve greater reliability.
• The MIMO channel between the transmitting entity and the receiving entity may experience various deleterious channel conditions such as, e.g., fading, multipath, and interference effects. In general, good performance may be achieved for data transmission via the MIMO channel if the interference and noise observed at the receiving entity are spatially "white", which is flat or constant interference and noise power across spatial dimension. This may not be the case, however, if the interference is from interfering sources located in specific directions. If the interference is spatially "colored" (not white), then the receiving entity can ascertain the spatial characteristics of the interference and place beam nulls in the direction of the interfering sources. The receiving entity may also provide the transmitting entity with channel state information (CSI). The transmitting entity can then spatially process data in a manner to maximize signal-to-noise-and-interference ratio (SNR) at the receiving entity. Good performance can thus be achieved when the transmitting and receiving entities perform the appropriate transmit and receive spatial processing for the data transmission in the presence of spatially colored interference.
• To perform spatial nulling of interference, the receiving entity typically needs to ascertain the characteristics of the interference. If the interference characteristics change over time, then the receiving entity would need to continually obtain up-to-date interference information in order to accurately place the beam nulls. The receiving entity may also need to continually send channel state information at a sufficient rate to allow the transmitting entity to perform the appropriate spatial processing. The need for accurate interference information and channel state information renders spatial nulling of interference not practical for most MIMO systems.
• XP010803892 "Multistream space-time coding by spatial spreading, scrambling and delay diversity" (Medles A et al) discloses a flexible spatial spreading and scrambling framework which allows to transmit an arbitrary number of streams. XP002357231 "Single-and Multi-Carrier DS-CDMA" (Hanzo L et al) discloses a GA-assisted scheme as a suboptimal multiuser detection technique in bit-synchronous CDMA systems over single-path Rayleigh fading channels. XP010638423 "Space-time spreading codecs for a multiuser mimo system" (Doostnejad R et al) discloses the use of the spatial dimension in enabling multiple access for the down-link of a cellular communication system and proposes a two dimensional spreading code.
• There is therefore a need in the art for techniques to transmit data in the presence of spatially colored interference and noise.
SUMMARY
• According to the present invention, a method for transmitting data from a transmitting entity to a receiving entity in a wireless multiple-input multiple-output (MIMO) communication system according to claim 1 and an apparatus in a wireless multiple-input multiple-output (MIMO) communication system according to claim 17 are provided.
• The present invention also provides a method for receiving a data transmission sent by a transmitting entity to a receiving entity in a wireless multiple-input multiple-output (MIMO) communication system according to claim 28 and an apparatus in a wireless multiple-input multiple-output (MIMO) communication system according to claim 42.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 shows a MIMO system with a transmitting entity, a receiving entity, and two interfering sources.
• FIG. 2 shows a model for data transmission with spatial spreading.
• FIG. 3 shows the processing performed by the transmitting entity.
• FIG. 4 shows the processing performed by the receiving entity.
• FIG. 5 shows a block diagram of the transmitting and receiving entities.
• FIG. 6 shows a transmit (TX) data processor and a TX spatial processor at the transmitting entity.
• FIG. 7 shows a receive (RX) spatial processor and an RX data processor at the receiving entity.
• FIG. 8 shows an RX spatial processor and an RX data processor that implement a successive interference cancellation (SIC) technique.
DETAILED DESCRIPTION
• The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
• Techniques for transmitting data with spatial spreading in single-carrier and multi-carrier MIMO systems are described herein. Spatial spreading refers to the transmission of a data symbol (which is a modulation symbol for data) on multiple eigenmodes or spatial channels (described below) of a MIMO channel simultaneously with a steering vector. The spatial spreading randomizes a transmission channel observed by a stream of data symbols, which effectively whitens the transmitted data symbol stream and can provide various benefits as described below.
• For data transmission with spatial spreading, a transmitting entity processes (e.g., encodes, interleaves, and modulates) each data packet to obtain a corresponding block of data symbols and multiplexes data symbol blocks onto N_S data symbol streams for transmission on N_S transmission channels in a MIMO channel. The transmitting entity then spatially spreads the N_S data symbol streams with steering matrices to obtain N_S spread symbol streams. The transmitting entity further spatially processes the N_S spread symbol streams for either full-CSI transmission on N_S eigenmodes of the MIMO channel or partial-CSI transmission on N_S spatial channels of the MIMO channel, as described below.
• A receiving entity obtains N_R received symbol streams via N_R receive antennas and performs receiver spatial processing for full-CSI or partial-CSI transmission to obtain N_S detected symbol streams, which are estimates of the N_S spread symbol streams. The receiving entity further spatially despreads the N_S detected symbol streams with the same steering matrices used by the transmitting entity and obtains N_S recovered symbol streams, which are estimates of the N_S data symbol streams. The receiver spatial processing and spatial despreading may be performed jointly or separately. The receiving entity then processes (e.g., demodulates, deinterleaves, and decodes) each block of recovered symbols in the N_S recovered symbol streams to obtain a corresponding decoded data packet.
• The receiving entity may also estimate the signal-to-noise-and-interference ratio (SNR) of each transmission channel used for data transmission and select a suitable rate for the transmission channel based on its SNR. The same or different rates may be selected for the N_S transmission channels. The transmitting entity encodes and modulates data for each transmission channel based on its selected rate.
• Various aspects and embodiments of the invention are described in further detail below.
• FIG. 1 shows a MIMO system 100 with a transmitting entity 110, a receiving entity 150, and two interfering sources 190a and 190b. Transmitting entity 110 transmits data to receiving entity 150 via line-of-sight paths (as shown in FIG. 1) and/or reflected paths (not shown in FIG. 1). Interfering sources 190a and 190b transmit signals that act as interference at receiving entity 150. The interference observed by receiving entity 150 from interfering sources 190a and 190b may be spatially colored.
1. Single-Carrier MIMO System
• For a single-carrier MIMO system, a MIMO channel formed by the N_T transmit antennas at the transmitting entity and the N_R receive antennas at the receiving entity may be characterized by an N_R × N_T channel response matrix H, which may be expressed as: $$\underline{\mathbf{H}}=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="imgf0006.png,imgf0007.png"\; state="\{\{state\}\}">h</figure-callout>}}_{1,1}& {\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="imgf0006.png,imgf0007.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="h"\; filenames="imgf0008.png"\; state="\{\{state\}\}">h</figure-callout>\; </figure-callout>}}_{1,2}& \cdots & {\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="imgf0006.png,imgf0007.png"\; state="\{\{state\}\}">h</figure-callout>}}_{\mathrm{1},{\mathrm{<figure-callout\; id="2"\; label="N\; T\; h"\; filenames="imgf0008.png"\; state="\{\{state\}\}">N</figure-callout>}}_{\mathrm{T}}}& \\ h_{}{}_{2,1}& {\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\; <figure-callout\; id="2"\; label="h"\; filenames="imgf0008.png"\; state="\{\{state\}\}">h</figure-callout>\; </figure-callout>}}_{2,2}& \cdots & {\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="imgf0008.png"\; state="\{\{state\}\}">h</figure-callout>}}_{\mathrm{2},{\mathrm{N}}_{\mathrm{T}}}\\ ⋮& ⋮& \ddots & ⋮\\ h_{{\mathrm{N}}_{\mathrm{R}},1}& h_{{\mathrm{N}}_{\mathrm{R}},2}& \cdots & h_{{\mathrm{N}}_{\mathrm{R}},{\mathrm{N}}_{\mathrm{T}}}\end{array}\right],$$

  

  where entry h_i,j, for i = 1 ... N_R and j = 1 ... N_T, denotes the coupling or complex channel gain between transmit antenna j and receive antenna i.
• Data may be transmitted in various manners in the MIMO system. For a full-CSI transmission scheme, data is transmitted on "eigenmodes" of the MIMO channel (described below). For a partial-CSI transmission scheme, data is transmitted on spatial channels of the MIMO channel (also described below).
A. Full-CSI Transmission
• For the full-CSI transmission scheme, eigenvalue decomposition may be performed on a correlation matrix of H to obtain N_S eigenmodes of H, as follows: $$\underline{\mathbf{R}}={\underline{\mathbf{H}}}^{\mathit{H}}\cdot \underline{\mathbf{H}}=\underline{\mathbf{E}}\cdot \underline{\mathbf{\Lambda }}\cdot {\underline{\mathbf{E}}}^{\mathit{H}},$$

  
• where R is an N_T × N_T correlation matrix of H;
• E is an N_T × N_T unitary matrix whose columns are eigenvectors of R;
• Λ is an N_T × N_T diagonal matrix of eigenvalues of R; and
• " ^H " denotes a conjugate transpose.
• A unitary matrix U is characterized by the property U ^H · U = I, where I is the identity matrix. The columns of a unitary matrix are orthogonal to one another.
• The transmitting entity may perform spatial processing with the eigenvectors of R to transmit data on the N_S eigenmodes of H. The eigenmodes may be viewed as orthogonal spatial channels obtained through decomposition. The diagonal entries of Λ are eigenvalues of R, which represent the power gains for the N_S eigenmodes.
• The transmitting entity performs spatial processing for full-CSI transmission as follows: $$\underline{\mathbf{x}}=\underline{\mathbf{E}}\cdot \underline{\mathbf{s}},$$

  
• where s is an N_T × 1 vector with N_S non-zero entries for N_S data symbols to be transmitted simultaneously on the N_S spatial channels; and
• x is an N_T × 1 vector with N_T transmit symbols to be sent from the N_T transmit antennas.
• The received symbols at the receiving entity may be expressed as: $$\underline{\mathbf{r}}=\underline{\mathbf{H}}\cdot \underline{\mathbf{x}}+\underline{\mathbf{j}},$$

  
• where r is an N_R × 1 vector with N_R received symbols obtained via the N_R receive antennas; and
• j is an N_R × 1 vector of interference and noise observed at the receiving entity.
• The receiving entity performs spatial processing with an N_T × N_R spatial filter matrix M = Λ ^-1 ·E ^H ·H ^H for full-CSI transmission, as follows: $$\begin{array}{cc}\underline{\hat{\mathbf{s}}}& =\underline{\mathbf{M}}\cdot \underline{\mathbf{r}}\hfill \\ & ={\underline{\mathrm{\Lambda }}}^{-\mathbf{1}}\cdot {\underline{\mathbf{E}}}^{\mathit{H}}\cdot {\underline{\mathbf{H}}}^{\mathit{H}}\cdot \left(\underline{\mathbf{H}}\cdot \underline{\mathbf{E}}\cdot \underline{\mathbf{s}}+\underline{\mathbf{j}}\right)\hfill \\ & ={\underline{\mathrm{\Lambda }}}^{-\mathbf{1}}\cdot {\underline{\mathbf{E}}}^{\mathit{H}}\cdot \underline{\mathbf{E}}\cdot \underline{\mathrm{\Lambda }}\cdot {\underline{\mathbf{E}}}^{\mathit{H}}\cdot \underline{\mathbf{E}}\cdot \underline{\mathbf{s}}+{\underline{\mathrm{\Lambda }}}^{-\mathbf{1}}\cdot {\underline{\mathbf{E}}}^{\mathit{H}}\cdot {\underline{\mathbf{H}}}^{\mathit{H}}\cdot \underline{\mathbf{j}}\hfill \\ & =\underline{\mathbf{s}}+\underline{\tilde{\mathbf{j}}}\hfill \end{array}$$

  
• where ŝ is an N_T × 1 vector with N_S recovered symbols or data symbol estimates, which are estimates of the N_S data symbols in s ; and
• j̃ = Λ ^-1 ·E ^H ·H ^H ·j is the "post-detection" interference and noise after the spatial processing at the receiving entity.
• An eigenmode may be viewed as an effective channel between an element of s and a corresponding element of ŝ with the transmitting and receiving entities performing the spatial processing shown in equations (3) and (5), respectively. The transmitting and receiving entities typically only have estimates of the channel response matrix H , which may be obtained based on pilot symbols. A pilot symbol is a modulation symbol for pilot, which is data that is known a priori by both the transmitting and receiving entities. For simplicity, the description herein assumes no channel estimation error.
• The vector j may be decomposed into an interference vector i and a noise vector n , as follows: $$\underline{\mathbf{j}}=\underline{\mathbf{i}}+\underline{\mathbf{n}}\mathbf{.}$$

  
• The noise may be characterized by an N_R × N_R autocovariance matrix ϕ _nn = E[ n · n ^H ], where E[x] is the expected value of x. If the noise is additive white Gaussian noise (AWGN) with zero mean and a variance of ${\mathrm{<figure-callout\; id="2"\; label="\sigma \; n"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}_n^{},$

  

  then the noise autocovariance matrix may be expressed as: ${\underline{\varphi }}_{\mathit{nn}}={\mathrm{<figure-callout\; id="2"\; label="\sigma \; n"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}_n^{}\cdot \underline{\mathbf{I}}\mathrm{.}$

  

  Similarly, the interference may be characterized by an N_R × N_R autocovariance matrix ϕ _ii = E[ i · i ^H ]. The autocovariance matrix of j may be expressed as ϕ _ii = E[ j·j ^H ]= ϕ _nn + ϕ _ii , assuming the interference and noise are uncorrelated.
• The interference and noise are considered to be spatially white if their autocovariance matrices are of the form σ ²· I due to the noise and interference being uncorrelated. For spatially white interference and noise, each receive antenna observes the same amount of interference and noise, and the interference and noise observed at each receive antenna are uncorrelated with the interference and noise observed at all other receive antennas. For spatially colored interference and noise, the autocovariance matrices have non-zero off-diagonal terms due to correlation between the interference and noise observed at different receive antennas. In this case, each receive antenna i may observe a different amount of interference and noise, which is equal to the sum of the N_R elements in the i-th row of the matrix ϕ _jj .
• If the interference and noise are spatially colored, then the optimal eigenvectors for full-CSI transmission may be derived as: $${\underline{\mathbf{R}}}_{\mathrm{opt}}={\underline{\mathbf{H}}}^{\mathit{H}}\cdot {\underline{\mathit{\varphi }}}_{\mathit{jj}}^{-\mathit{1}}\cdot \underline{\mathbf{H}}={\underline{\mathbf{E}}}_{\mathrm{opt}}\cdot \underline{\mathrm{\Lambda }}\cdot {\underline{\mathbf{E}}}_{\mathrm{opt}}^{\mathit{H}}\mathbf{.}$$

  
• The eigenvectors E _opt steer the data transmission in the direction of the receiving entity and further place beam nulls in the direction of the interference. However, the transmitting entity would need to be provided with the autocovariance matrix ϕ _jj in order to derive the eigenvectors E _opt. The matrix ϕ _jj is based on the interference and noise observed at the receiving entity and can only be determined by the receiving entity. To spatially null the interference, the receiving entity would need to send this matrix, or its equivalent, back to the transmitting entity, which can represent a large amount of channel state information to send back.
• Spatial spreading may be used to spatially whiten the interference and noise observed by the receiving entity and may potentially improve performance. The transmitting entity performs spatial spreading with an ensemble of steering matrices such that the complementary spatial despreading at the receiving entity spatially whitens the interference and noise.
• For full-CSI transmission with spatial spreading, the transmitting entity performs processing as follows: $${\underline{\mathbf{x}}}_{\mathit{fcsi}}\left(m\right)=\underline{\mathbf{E}}\left(m\right)\cdot \underline{\mathbf{V}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right),$$

  
• where s (m) is a data symbol vector for transmission span m;
• V (m) is an N_T × N_T steering matrix for transmission span m;
• E (m) is a matrix of eigenvectors for transmission span m; and
• x _fcsi (m) is a transmit symbol vector for transmission span m.
• A transmission span may cover time and/or frequency dimensions. For example, in a single-carrier MIMO system, a transmission span may correspond to one symbol period, which is the time duration to transmit one data symbol. A transmission span may also cover multiple symbol periods. As shown in equation (8), each data symbol in s (m) is spatially spread with a respective column of V (m) to obtain N_T spread symbols, which may then be transmitted on all eigenmodes of H (m).
• The received symbols at the receiving entity may be expressed as: $${\underline{\mathbf{r}}}_{\mathit{fsci}}\left(m\right)=\underline{\mathbf{H}}\left(m\right)\cdot {\underline{\mathbf{x}}}_{\mathit{fcsi}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)=\underline{\mathbf{H}}\left(m\right)\cdot \underline{\mathbf{E}}\left(m\right)\cdot \underline{\mathbf{V}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)\mathrm{.}$$

  
• The receiving entity derives a spatial filter matrix M _fcsi (m) as follows: $${\underline{\mathbf{M}}}_{\mathit{fsci}}\left(m\right)={\underline{\mathrm{\Lambda }}}^{-1}\left(m\right)\cdot {\underline{\mathbf{E}}}^H\left(m\right)\cdot {\underline{\mathbf{H}}}^H\left(m\right)\mathrm{.}$$

  
• The receiving entity performs receiver spatial processing and spatial despreading using M _fcsi (m) and V ^H (m), respectively, as follows: $$\begin{array}{cc}{\underline{\hat{\mathbf{s}}}}_{\mathit{fcsi}}\left(m\right)& ={\underline{\mathbf{V}}}^H\left(m\right)\cdot {\underline{\mathbf{M}}}_{\mathit{fcsi}}\left(m\right)\cdot {\underline{\mathbf{r}}}_{\mathit{fcsi}}\left(m\right),\hfill \\ & ={\underline{\mathbf{V}}}^H\left(m\right)\cdot {\underline{\mathrm{\Lambda }}}^{-1}\left(m\right)\cdot {\underline{\mathbf{E}}}^H\left(m\right)\cdot {\underline{\mathbf{H}}}^H\left(m\right)\cdot \left[\underline{\mathbf{H}}\left(m\right)\cdot \underline{\mathbf{E}}\left(m\right)\cdot \underline{\mathbf{V}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)\right],\hfill \\ & =\underline{\mathbf{s}}\left(m\right)+{\underline{\mathbf{j}}}_{\mathit{fcsi}}\left(m\right),\hfill \end{array}$$

  

  where j _fcsi (m) is the "post-detection" interference and noise after the spatial processing and spatial despreading at the receiving entity, which is: $${\underline{\mathbf{j}}}_{\mathrm{fcsi}}\left(m\right)={\underline{\mathbf{V}}}^H\left(m\right)\cdot {\underline{\mathrm{\Lambda }}}^{-1}\left(m\right)\cdot {\underline{\mathbf{E}}}^H\left(m\right)\cdot {\underline{\mathbf{H}}}^H\left(m\right)\cdot \underline{\mathbf{j}}\left(m\right)\mathrm{.}$$

  
• As shown in equation (12), the received interference and noise in j(m) are transformed by the conjugate transposes of V (m), E (m), and H (m). E (m) is a matrix of eigenvectors that may not be optimally computed for spatially colored interference and noise if the autocovariance matrix ϕ _jj (m) is not known, which is often the case.
• The transmitting and receiving entities may, by random chance, operate with a matrix E (m) that results in more interference and noise being observed by the receiving entity. This may be the case, for example, if a mode of E (m) is correlated with the interference. If the MIMO channel is static, then the transmitting and receiving entities may continually operate with a matrix E (m) that provides poor performance. The spatial despreading with the steering matrix V (m) spatially whitens the interference and noise. The effectiveness of the interference and noise whitening is dependent on the characteristics of the channel response matrix H (m) and the interference j (m). If a high degree of correlation exists between the desired signal and the interference, then this limits the amount of gain provided by the whitening of the interference and noise.
• The SNR of each eigenmode with full-CSI transmission may be expressed as: $${\mathit{\gamma }}_{\mathit{fcsi},\ell }\left(k\right)=\frac{P_{\ell }\left(m\right){\lambda }_{\ell }\left(m\right)}{{\sigma }_{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">j</figure-callout>}}^2},\mathrm{for}\ell =1\dots N_s,$$

  
• where P_ℓ (m) is the transmit power used for the transmit symbol sent on eigeninode ℓ in transmission span m;
• λ_ℓ (m) is the eigenvalue for eigenmode ℓ in transmission span m, which is the ℓ-th diagonal element of Λ(m);
• ${\mathrm{<figure-callout\; id="2"\; label="\sigma \; j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}_j^{}$
  
  is the variance of the received interference and noise; and
• γ_fcsi,ℓ (m) is the SNR of eigenmode ℓ in transmission span m.
B. Partial-CSI Transmission
• For partial-CSI transmission with spatial spreading, the transmitting entity performs processing as follows: $${\underline{\mathbf{x}}}_{\mathit{pcsi}}\left(m\right)=\underline{\mathbf{V}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right),$$

  

  where x _pcsi (m) is the transmit data vector for transmission span m. As shown in equation (14), each data symbol in s (m) is spatially spread with a respective column of V (m) to obtain N_T spread symbols, which may then be transmitted from all N_T transmit antennas.
• The received symbols at the receiving entity may be expressed as: $${\underline{\mathbf{r}}}_{\mathit{pcsi}}\left(m\right)=\underline{\mathbf{H}}\left(m\right)\cdot \underline{\mathbf{V}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)={\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right),$$

  
• where r _pcsi (m) is the received symbol vector for transmission span m; and
• H _eff (m) is an effective channel response matrix, which is:
$${\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)=\underline{\mathbf{H}}\left(m\right)\cdot \underline{\mathbf{V}}\left(m\right)\mathrm{.}$$

• The receiving entity may derive estimates of the transmitted data symbols in s using various receiver processing techniques. These techniques include a channel correlation matrix inversion (CCMI) technique (which is also commonly referred to as a zero-forcing technique), a minimum mean square error (MMSE) technique, a successive interference cancellation (SIC) technique, and so on. The receiving entity may perform receiver spatial processing and spatial despreading jointly or separately, as described below. In the following description, one data symbol stream is sent for each element of the data symbol vector s .
• For the CCMI technique, the receiving entity may derive a spatial filter matrix M _ccmi (m), as follows: $${\underline{\mathbf{M}}}_{\mathit{ccmi}}\left(m\right)={\left[{\underline{\mathbf{H}}}_{\mathit{eff}}^H\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)\right]}^{-1}\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)={\underline{\mathbf{R}}}_{\mathit{eff}}^{-1}\left(m\right)\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\mathrm{.}$$

  
• The receiving entity may then perform CCMI spatial processing and despreading jointly, as follows: $$\begin{array}{cc}{\underline{\hat{\mathbf{s}}}}_{\mathit{ccmi}}\left(m\right)& ={\underline{\mathbf{M}}}_{\mathit{ccmi}}\left(m\right)\cdot {\underline{\mathbf{r}}}_{\mathit{pcsi}}\left(m\right),\hfill \\ & ={\underline{\mathbf{R}}}_{\mathit{eff}}^{-1}\left(m\right)\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\cdot \left[{\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)\right],\hfill \\ & =\underline{\mathbf{s}}\left(m\right)+{\underline{\mathbf{j}}}_{\mathit{ccmi}}\left(m\right),\hfill \end{array}$$

  

  where j _ccmi (m) is the CCMI filtered and despread interference and noise, which is: $${\underline{\mathbf{j}}}_{\mathit{ccmi}}\left(m\right)={\underline{\mathbf{R}}}_{\mathit{eff}}^{-1}\left(m\right)\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\cdot \underline{\mathbf{j}}\left(m\right)={\underline{\mathbf{V}}}^H\left(m\right)\cdot {\underline{\mathbf{R}}}^{-1}\left(m\right)\cdot {\underline{\mathbf{H}}}^H\left(m\right)\cdot \underline{\mathbf{j}}\left(m\right)\mathrm{.}$$

  
• As shown in equation (19), the interference and noise j (m) is whitened by V ^H (m). However, due to the structure of R (m), the CCMI technique may amplify the interference and noise.
• The receiving entity may also perform CCMI spatial processing and spatial despreading separately, as follows: $$\begin{array}{cc}{\underline{\hat{\mathbf{s}}}}_{\mathit{ccmi}}\left(m\right)& ={\underline{\mathbf{V}}}^H\left(m\right)\cdot {\underline{\tilde{\mathbf{M}}}}_{\mathit{ccmi}}\left(m\right)\cdot {\underline{\mathbf{r}}}_{\mathit{pcsi}}\left(m\right),\hfill \\ & ={\underline{\mathbf{V}}}^H\left(m\right)\cdot {\underline{\mathbf{R}}}^{-1}\left(m\right)\cdot {\underline{\mathbf{H}}}^H\left(m\right)\cdot \left[\underline{\mathbf{H}}\left(m\right)\cdot \underline{\mathbf{H}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)\right],\hfill \\ & =\underline{\mathbf{s}}\left(m\right)+{\underline{\mathbf{j}}}_{\mathit{ccmi}}\left(m\right),\hfill \end{array}$$

  

  where M̃ _ccmi (m) = R ^-1(m)· H ^H (m). In any case, a spatial channel may be viewed as an effective channel between an element of s and a corresponding element of ŝ with the transmitting entity performing spatial processing with the identity matrix I and the receiving entity performing the appropriate receiver spatial processing to estimate s .
• The SNR for the CCMI technique may be expressed as: $${\mathit{\gamma }}_{\mathit{ccmi},\ell }\left(m\right)=\frac{P_{\ell }\left(m\right)}{r_{\ell \ell }\left(m\right){\sigma }_{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">j</figure-callout>}}^2},\mathrm{for}\ell =1\dots N_{\mathrm{s}},$$

  
• where P_ℓ (m) is the power used for data symbol stream {s_ℓ } in transmission span m;
• r_ℓℓ (m) is the ℓ-th diagonal element of ${\underline{\mathbf{R}}}_{\mathit{eff}}^{-1}\left(m\right);$
  
• ${\mathrm{<figure-callout\; id="2"\; label="\sigma \; j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}_j^{}$
  
  is the variance of the received interference and noise; and
• γ_ccmi,ℓ (m) is the SNR of data symbol stream {s _ℓ} in transmission span m.
• The quantity $P_{\ell }\left(m\right)/{\mathrm{<figure-callout\; id="2"\; label="\sigma \; j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}_j^{}$

  

  is the SNR of data symbol stream {s _ℓ} at the receiving entity prior to the receiver spatial processing and is commonly referred to as the received SNR. The quantity γ_ccmi,ℓ (m) is the SNR of data symbol stream {s _ℓ} after the receiver spatial processing and is also referred to as the post-detection SNR. In the following description, "SNR" refers to post-detection SNR unless noted otherwise.
• For the MMSE technique, the receiving entity may derive a spatial filter matrix M _mmse (m), as follows: $${\underline{\mathbf{M}}}_{\mathit{mmse}}\left(m\right)={\left[{\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)+{\underline{\varphi }}_{\mathit{jj}}\left(m\right)\right]}^{-1}\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\mathrm{.}$$

  
• The spatial filter matrix M _mmse (m) minimizes the mean square error between the symbol estimates from the spatial filter and the data symbols. If the autocovariance matrix ϕ _jj (m) is not known, which is often the case, then the spatial filter matrix M _mmse (m) maybe approximated as: $${\underline{\mathbf{M}}}_{\mathit{mmse}}\left(m\right)={\left[{\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)+{\mathrm{<figure-callout\; id="2"\; label="\sigma \; j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}_j^{}\cdot \underline{\mathbf{I}}\right]}^{-1}\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\mathrm{.}$$

  
• The receiving entity may perform MMSE spatial processing and despreading jointly, as follows: $$\begin{array}{cc}{\underline{\hat{\mathbf{s}}}}_{\mathit{mmse}}\left(m\right)& ={\underline{\mathbf{D}}}_{\mathrm{Q}}\left(m\right)\cdot {\underline{\mathbf{M}}}_{\mathit{mmse}}\left(m\right)\cdot {\underline{\mathbf{r}}}_{\mathit{pcsi}}\left(m\right),\hfill \\ & ={\underline{\mathbf{D}}}_{\mathrm{Q}}\left(m\right)\cdot {\underline{\mathbf{M}}}_{\mathit{mmse}}\left(m\right)\cdot \left[{\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+\underline{\mathbf{j}}\left(m\right)\right],\hfill \\ & ={\underline{\mathbf{D}}}_{\mathrm{Q}}\left(m\right)\cdot \underline{\mathbf{Q}}\left(m\right)\cdot \underline{\mathbf{s}}\left(m\right)+{\underline{\mathbf{j}}}_{\mathit{mmse}}\left(m\right),\hfill \end{array}$$

  
• where Q (m) = M _mmse (m)· H _eff (m);
• D _Q(m) is a diagonal matrix whose diagonal elements are the diagonal elements
• of Q ^-1(m), or D _Q(m)=[diag [ Q (m)]]^-1; and
• j _mmse (m) is the MMSE filtered and despread interference and noise, which is:
$$\begin{array}{cc}{\underline{\mathbf{j}}}_{\mathit{mmse}}\left(m\right)& ={\underline{\mathbf{D}}}_{\mathrm{Q}}\left(m\right)\cdot {\underline{\mathbf{M}}}_{\mathit{mmse}}\left(m\right)\cdot \underline{\mathbf{j}}\left(m\right),\hfill \\ & ={\underline{\mathbf{D}}}_{\mathrm{Q}}\left(m\right)\cdot {\left[{\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}\left(m\right)+{\underline{\mathit{\varphi }}}_{\mathit{jj}}\left(m\right)\right]}^{-1}\cdot {\underline{\mathbf{H}}}_{\mathit{eff}}^H\left(m\right)\cdot \underline{\mathbf{j}}\left(m\right)\mathrm{.}\hfill \end{array}$$

• The symbol estimates from the spatial filter matrix M _mmse (m) are unnormalized estimates of the data symbols. The multiplication with D _Q(m) provides normalized estimates of the data symbols. The receiving entity may also perform MMSE spatial processing and spatial despreading separately, similar to that described above for the CCMI technique.
• The SNR for the MMSE technique may be expressed as: $${\mathit{\gamma }}_{\mathit{mmse},\ell }\left(m\right)=\frac{q_{\ell \ell }\left(m\right)}{1-q_{\ell \ell }\left(m\right)}{P}_{\ell }\left(m\right),\mathrm{for}\ell =1\dots N_{\mathrm{S}},$$

  
• where q_ℓℓ (m) is the ℓ-th diagonal element of Q (m); and
• γ_mmse,ℓ (m) is the SNR of data symbol stream {s _ℓ} in transmission span m.
• For the SIC technique, the receiving entity processes the N_R received symbol streams in N_S successive stages to recover the N_S data symbol streams. For each stage ℓ, the receiving entity performs spatial processing and despreading on either the N_R received symbol streams or N_R modified symbol streams from the preceding stage (e.g., using the CCMI, MMSE, or some other technique) to obtain one recovered symbol stream {ŝ _ℓ}. The receiving entity then processes (e.g., demodulates, deinterleaves, and decodes) this recovered symbol stream to obtain a corresponding decoded data stream {d̂ _ℓ}. The receiving entity next estimates the interference this stream causes to the other data symbol streams not yet recovered. To estimate the interference, the receiving entity re-encodes, interleaves, and symbol maps the decoded data stream in the same manner performed at the transmitting entity for this stream and obtains a stream of "remodulated" symbols {

  

  }, which is an estimate of the data symbol stream just recovered. The receiving entity then spatially spreads the remodulated symbol stream with the steering matrix V (m) and further multiplies the result with the channel response matrix H (m) for each transmission span of interest to obtain N_R interference components caused by this stream. The N_R interference components are then subtracted from the N_R modified or received symbol streams for the current stage to obtain N_R modified symbol streams for the next stage. The receiving entity then repeats the same processing on the N_R modified symbol streams to recover another data stream.
• For the SIC technique, the SNR of each data symbol stream is dependent on (1) the spatial processing technique (e.g., CCMI or MMSE) used for each stage, (2) the specific stage in which the data symbol stream is recovered, and (3) the amount of interference due to the data symbol streams not yet recovered. In general, the SNR progressively improves for data symbol streams recovered in later stages because the interference from data symbol streams recovered in prior stages is canceled. This then allows higher rates to be used for data symbol streams recovered in later stages.
C. System Model
• FIG. 2 shows a model for data transmission with spatial spreading. Transmitting entity 110 performs spatial spreading (block 220) and spatial processing for full-CSI or partial-CSI transmission (block 230). Receiving entity 150 performs receiver spatial processing for full-CSI or partial-CSI transmission (block 260) and spatial despreading (block 270). The description below makes references to the vectors shown in FIG. 2.
• FIG. 3 shows a process 300 performed by the transmitting entity to transmit data with spatial spreading in the MIMO system. The transmitting entity processes (e.g., encodes and interleaves) each packet of data to obtain a corresponding block of coded data, which is also called a code block or a coded data packet (block 312). Each code block is encoded separately at the transmitting entity and decoded separately at the receiving entity. The transmitting entity further symbol maps each code block to obtain a corresponding block of data symbols (also block 312). The transmitting entity multiplexes all data symbol blocks generated for all data packets onto N_S data symbol streams (denoted by vector s ) (block 314). Each data symbol stream is sent on a respective transmission channel. The transmitting entity spatially spreads the N_S data symbol streams with steering matrices and obtains N_S spread symbol streams (denoted by a vector w in FIG. 2) (block 316). The spatial spreading is such that each data symbol block is spatially spread with multiple (N_M) steering matrices to randomize the transmission channel observed by the block. The randomization of the transmission channel results from using different steering matrices and not necessarily from randomness in the elements of the steering matrices. The transmitting entity further performs spatial processing on the N_S spread symbol streams for full-CSI or partial-CSI transmission, as described above, and obtains N_T transmit symbol streams (denoted by vector x ) (block 318). The transmitting entity then conditions and sends the N_T transmit symbol streams via the N_T transmit antennas to the receiving entity (block 320).
• FIG. 4 shows a process 400 performed by the receiving entity to receive data transmitted with spatial spreading in the MIMO system. The receiving entity obtains N_R received symbol streams (denoted by vector r ) via the N_R receive antennas (block 412). The receiving entity estimates the response of the MIMO channel (block 414), performs spatial processing for full-CSI or partial-CSI transmission based on the MIMO channel estimate, and obtains N_S detected symbol streams (denoted by a vector ŵ in FIG. 2) (block 416). The receiving entity further spatially despreads the N_S detected symbol streams with the same steering matrices used by the transmitting entity and obtains N_S recovered symbol streams (denoted by vector ŝ ) (block 418). The receiver spatial processing and spatial despreading may be performed jointly or separately, as described above. The receiving entity then processes (e.g., demodulates deinterleaves, and decodes) each block of recovered symbols in the N_S recovered symbol streams to obtain a corresponding decoded data packet (block 420). The receiving entity may also estimate the SNR of each transmission channel used for data transmission and select a suitable rate for the transmission channel based on its SNR (block 422). The same or different rates may be selected for the N_S transmission channels.
• Referring back to FIG. 2, the N_S data symbol streams are sent on N_S transmission channels of the MIMO channel. Each transmission channel is an effective channel observed by a data symbol stream between an element of the vector s at the transmitting entity and a corresponding element of the vector ŝ at the receiving entity (e.g., the ℓ-th transmission channel is the effective channel between the ℓ-th element of s and the ℓ-th element of ŝ ). The spatial spreading randomizes the N_S transmission channels. The N_S spread symbol streams are sent on either the N_S eigenmodes of the MIMO channel for full-CSI transmission or the N_S spatial channels of the MIMO channel for partial-CSI transmission.
D. Spatial Spreading
• The steering matrices used for spatial spreading may be generated in various manners, as described below. In one embodiment, a set of L steering matrices is generated and denoted as { V }, or V (i) for i = 1 ... L, where L may be any integer greater than one. These steering matrices are unitary matrices having orthogonal columns. Steering matrices from this set are selected and used for spatial spreading.
• The spatial spreading may be performed in various manners. In general, it is desirable to use as many different steering matrices as possible for each data symbol block so that the interference and noise are randomized across the block. Each data symbol block is transmitted in N_M transmission spans, where N_M > 1, and N_M is also referred to as the block length. One steering matrix in the set may be used for each transmission span. The transmitting and receiving entities may be synchronized such that both entities know which steering matrix to use for each transmission span. With spatial spreading, the receiving entity observes a distribution of interference and noise across each data symbol block even if the MIMO channel is constant across the entire block. This avoids the case in which high levels of interference and noise are received because the transmitting and receiving entities continually use a bad matrix of eigenvectors or the receiving entity continually observes colored interference.
• The L steering matrices in the set may be selected for use in various manners. In one embodiment, the steering matrices are selected from the set in a deterministic manner. For example, the L steering matrices may be cycled through and selected in sequential order, starting with the first steering matrix V (1), then the second steering matrix V (2), and so on, and then the last steering matrix V (L). In another embodiment, the steering matrices are selected from the set in a pseudo-random manner. For example, the steering matrix to use for each transmission span m may be selected based on a function f(m) that pseudo-randomly selects one of the L steering matrices, or steering matrix V (f(m)). In yet another embodiment, the steering matrices are selected from the set in a "permutated" manner. For example, the L steering matrices may be cycled through and selected for use in sequential order. However, the starting steering matrix for each cycle may be selected in a pseudo-random manner, instead of always being the first steering matrix V (1). The L steering matrices may also be selected in other manners, and this is within the scope of the invention.
• The steering matrix selection may also be dependent on the number of steering matrices (L) in the set and the block length (N_M). In general, the number of steering matrices may be greater than, equal to, or less than the block length. Steering matrix selection for these three cases may be performed as described below.
• If L = N_M, then the number of steering matrices matches the block length. In this case, a different steering matrix may be selected for each of the N_M transmission spans used to send each data symbol block. The N_M steering matrices for the N_M transmission spans may be selected in a deterministic, pseudo-random, or permutated manner, as described above.
• If L < N_M, then the block length is longer than the number of steering matrices in the set. In this case, the steering matrices are reused for each data symbol block and may be selected as described above.
• If L > N_M, then a subset of the steering matrices is used for each data symbol block. The selection of the specific subset to use for each data symbol block may be deterministic or pseudo-random. For example, the first steering matrix to use for the current data symbol block may be the steering matrix after the last one used for a prior data symbol block.
• As noted above, a transmission span may cover one or multiple symbol periods and/or one or multiple subbands. For improved performance, it is desirable to select the transmission span to be as small as possible so that (1) more steering matrices can be used for each data symbol block and (2) each receiving entity can obtain as many "looks" of the MIMO channel as possible for each data symbol block. The transmission span should also be shorter than the coherence time of the MIMO channel, which is the time duration over which the MIMO channel can be assumed to be approximately static. Similarly, the transmission span should be smaller than the coherence bandwidth of the MIMO channel for a wideband system (e.g., an OFDM system).
E. Applications for Spatial Spreading
• Spatial spreading may be used to randomize and whiten spatially colored interference and noise for both full-CSI and partial-CSI transmission, as described above. This may improve performance for certain channel conditions.
• Spatial spreading may also be used to reduce outage probability under certain operating scenarios. As an example, a block of data symbols for a code block may be partitioned into N_T data symbol subblocks. Each data symbol subblock may be coded and modulated based on the SNR expected for the subblock. Each data symbol subblock may be transmitted as one element of the data symbol vector s , and the N_T data symbol subblocks may be transmitted in parallel. An outage may then occur if any one of the N_T data symbol subblocks cannot be decoded error free by the receiving entity.
• If partial-CSI transmission without spatial spreading is used for the N_T data symbol subblocks, then each subblock is transmitted from a respective transmit antenna. Each data symbol subblock would then observe the SNR achieved for the spatial channel corresponding to its transmit antenna. The receiving entity can estimate the SNR of each spatial channel, select an appropriate rate for each spatial channel based on its SNR, and provide the rates for all N_T spatial channels to the transmitting entity. The transmitting entity can then encode and modulate the N_T data symbol subblocks based on their selected rates.
• The MIMO channel may change between time n when the rates are selected to time n+τ when the rates are actually used. This may be the case, for example, if the receiving entity has moved to a new location, if the MIMO channel changes faster than the feedback rate, and so on. The new channel response matrix H ₁ at time n+τ may have the same capacity as the prior channel response matrix H ₀ at time n, which may be expressed as: $$\mathrm{Cap}\left({\underline{\mathbf{H}}}_{\mathrm{0}}\right)={\displaystyle \sum _{\mathit{i}=\mathrm{1}}^{{\mathrm{<figure-callout\; id="2"\; label="N\; T\; log"\; filenames="imgf0008.png"\; state="\{\{state\}\}">N</figure-callout>}}_{\mathrm{T}}}}{\log }_{}{\mathrm{}}_{\mathrm{2}}\left(\mathrm{1}+{\mathit{\gamma }}_{\mathit{i}}\left(\mathit{n}\right)\right)={\displaystyle \sum _{\mathrm{i}=\mathrm{1}}^{{\mathrm{<figure-callout\; id="2"\; label="N\; T\; log"\; filenames="imgf0008.png"\; state="\{\{state\}\}">N</figure-callout>}}_{\mathrm{T}}}}{\log }_{}{\mathrm{}}_{\mathrm{2}}\left(\mathrm{1}+{\mathit{\gamma }}_{\mathit{i}}\left(\mathit{n}+\mathit{\tau }\right)\right)=\mathrm{Cap}\left({\underline{\mathbf{H}}}_{\mathrm{1}}\right),$$

  

  where γ_i (n) is the SNR of spatial channel i at time n and log₂(1+γ_i (n)) is the capacity of spatial channel i at time n. Even if the capacities of H ₀ and H ₁ are the same, the capacities of the individual spatial channels may have changed between time n and time n+τ, so that γ_i (n) may not be equal to γ_i (n+τ).
• Without spatial spreading, the outage probability increases if γ_i (n)< γ_i (n+τ) for any spatial channel i. This is because a data symbol subblock sent on a spatial channel with a lower SNR is less likely to be decoded error free, and any data symbol subblock decoded in error corrupts the entire data symbol block under the above assumption.
• If partial-CSI transmission with spatial spreading is used for the N_T data symbol subblocks, then each subblock is spatially spread and transmitted from all N_T transmit antennas. Each data symbol subblock would then be transmitted on a transmission channel formed by a combination of N_T spatial channels of the MIMO channel and would observe an effective SNR that is a combination of the SNRs for these spatial channels. The transmission channel for each data symbol subblock is determined by the steering matrices used for spatial spreading. If a sufficient number of steering matrices is used to spatially spread the N_T data symbol subblocks, then the effective SNR observed by each data symbol subblock will be approximately equal to the average SNR for all of the spatial channels when a powerful error correction code is employed. With spatial spreading, the outage probability may then be dependent on the average SNR of the spatial channels instead of the SNRs of the individual spatial channels. Thus, if the average SNR at time n+τ is approximately equal to the average SNR at time n, then the outage probability may be approximately the same even though the SNRs of the individual spatial channels may have changed between times n and n+τ.
• Spatial spreading can thus improve performance for the case in which inaccurate partial CSI is available at the transmitting entity and/or receiving entity. The inaccurate partial CSI may result from mobility, inadequate feedback rate, and so on.
2. Multi-Carrier MIMO System
• Spatial spreading may also be used for a multi-carrier MIMO system. Multiple carriers may be provided by orthogonal frequency division multiplexing (OFDM) or some other constructs. OFDM effectively partitions the overall system bandwidth into multiple (N_F) orthogonal frequency subbands, which are also referred to as tones, subcarriers, bins, and frequency channels. With OFDM, each subband is associated with a respective subcarrier that may be modulated with data. For an OFDM-based system, spatial spreading may be performed on each of the subbands used for data transmission.
• For a MIMO system that utilizes OFDM (i.e., a MIMO-OFDM system), one data symbol vector s (k,n) may be formed for each subband k in each OFDM symbol period n. Vector s (k,n) contains up to N_S data symbols to be sent via the N_S eigenmodes or spatial channels of subband k in OFDM symbol period n. Up to N_F vectors, s (k,n) for k = 1 ... N_F, may be transmitted concurrently on the N_F subbands in one OFDM symbol period. For the MIMO-OFDM system, a transmission span can cover both time and frequency dimensions. The index m for transmission span may thus be substituted with k,n for subband k and OFDM symbol period n. A transmission span may cover one subband in one OFDM symbol period or multiple OFDM symbol periods and/or multiple subbands.
• For the full-CSI transmission scheme, the channel response matrix H (k) for each subband k may be decomposed to obtain the N_S eigenmodes of that subband. The eigenvalues in each diagonal matrix Λ(k), for k = 1 ... N_F, may be ordered such that the first column contains the largest eigenvalue, the second column contains the next largest eigenvalue, and so on, or λ ₁(k) ≥ λ ₂(k) ≥ ... ≥ λ _NS (k), where λ _ℓ(k) is the eigenvalue in the ℓ-th column of Λ(k) after the ordering. When the eigenvalues for each matrix H (k) are ordered, the eigenvectors (or columns) of the associated matrix E (k) for that subband are also ordered correspondingly. A "wideband" eigenmode may be defined as the set of same-order eigenmodes of all N_F subbands after the ordering (e.g., the ℓ-th wideband eigenmode includes the ℓ-th eigenmode of all subbands). Each wideband eigenmode is associated with a respective set of N_F eigenvectors for the N_F subbands. The principle wideband eigenmode is the one associated with the largest eigenvalue in each matrix Λ(k) after the ordering. Data may be transmitted on the N_S wideband eigenmodes.
• For the partial-CSI transmission scheme, the transmitting entity may perform spatial spreading and spatial processing for each subband, and the receiving entity may perform receiver spatial processing and spatial despreading for each subband.
• Each data symbol block may be transmitted in various manners in the MIMO-OFDM system. For example, each data symbol block may be transmitted as one entry of the vector s (k,n) for each of the N_F subbands. In this case, each data symbol block is sent on all N_F subbands and achieves frequency diversity in combination with spatial diversity provided by spatial spreading. Each data symbol block may also span one or multiple OFDM symbol periods. Each data symbol block may thus span frequency and/or time dimensions (by system design) plus spatial dimension (with spatial spreading).
• The steering matrices may also be selected in various manners for the MIMO-OFDM system. The steering matrices for the subbands may be selected in a deterministic, pseudo-random, or permutated manner, as described above. For example, the L steering matrices in the set may be cycled through and selected in sequential order for subbands 1 through N_F in OFDM symbol period n, then subbands 1 through N_F in OFDM symbol period n+1, and so on. The number of steering matrices in the set may be less than, equal to, or greater than the number of subbands. The three cases described above for L = N_M, L < N_M, and L > N_M may also be applied for the subbands, with N_M being replaced with N_F.
3. MIMO System
• FIG. 5 shows a block diagram of transmitting entity 110 and receiving entity 150. At transmitting entity 110, a TX data processor 520 receives and processes (e.g., encodes, interleaves, and modulates) data and provides data symbols. A TX spatial processor 530 receives the data symbols, performs spatial spreading and spatial processing for full-CSI or partial-CSI transmission, multiplexes in pilot symbols, and provides N_T transmit symbol streams to N_T transmitter units (TMTR) 532a through 532t. Each transmitter unit 532 performs OFDM modulation (if applicable) and further conditions (e.g., converts to analog, filters, amplifies, and frequency upconverts) a respective transmit symbol stream to generate a modulated signal. N_T transmitter units 532a through 532t provide N_T modulated signals for transmission from N_T antennas 534a through 534t, respectively.
• At receiving entity 150, N_R antennas 552a through 552r receive the N_T transmitted signals, and each antenna 552 provides a received signal to a respective receiver unit (RCVR) 554. Each receiver unit 554 performs processing complementary to that performed by transmitter unit 532 (including OFDM demodulation, if applicable) and provides (1) received data symbols to an RX spatial processor 560 and (2) received pilot symbols to a channel estimator 584 within a controller 580. RX spatial processor 560 performs receiver spatial processing and spatial despreading on N_R received symbol streams from N_R receiver units 554 with spatial filter matrices and steering matrices, respectively, from controller 580 and provides N_S recovered symbol streams. An RX data processor 570 then processes (e.g., demaps, deinterleaves, and decodes) the recovered symbols and provides decoded data.
• Channel estimator 584 may derive Ĥ (m), which is an estimate of the channel response matrix H (m), based on pilot symbols transmitted without spatial spreading. Alternatively, channel estimator 584 may directly derive Ĥ _eff (m), which is an estimate of the effective channel response matrix H _eff (m), based on pilot symbols transmitted with spatial spreading. In any case, Ĥ (m) or Ĥ _eff (m) maybe used to derive the spatial filter matrix. Channel estimator 584 further estimates the SNR of each transmission channel based on received pilot symbols and/or received data symbols. The MIMO channel includes N_S transmission channels for each subband, but these transmission channels can be different depending on (1) whether full-CSI or partial-CSI transmission is used, (2) whether or not spatial spreading was performed, and (3) the specific spatial processing technique used by the receiving entity. Controller 580 selects a suitable rate for each transmission channel based on its SNR. Each selected rate is associated with a particular coding scheme and a particular modulation scheme, which collectively determine a data rate. The same or different rates may be selected for the N_S transmission channels.
• The rates for all transmission channels, other information, and traffic data are processed (e.g., encoded and modulated) by a TX data processor 590, spatially processed (if needed) by a TX spatial processor 592, conditioned by transmitter units 554a through 554r, and sent via antennas 552a through 552r. At transmitting entity 110, the N_R signals sent by receiving entity 150 are received by antennas 534a through 534t, conditioned by receiver units 532a through 532t, spatially processed by an RX spatial processor 544, and further processed (e.g., demodulated and decoded) by an RX data processor 546 to recover the selected rates. Controller 540 may then direct TX data processor 520 to process data for each transmission channel based on the rate selected for that transmission channel.
• Controllers 540 and 580 also control the operation of various processing units at transmitting entity 110 and receiving entity 150, respectively. Memory units 542 and 582 store data and/or program code used by controllers 540 and 580, respectively.
• FIG. 6 shows a block diagram of an embodiment of TX data processor 520 and TX spatial processor 530 at transmitting entity 110. For this embodiment, TX data processor 520 includes N_D TX data stream processors 620a through 620nd for N_D data streams {d _ℓ}, for ℓ = 1 ... N_D, where in general N_D ≥ 1.
• Within each TX data stream processor 620, an encoder 622 receives and encodes its data stream {d _ℓ} based on a coding scheme and provides code bits. Each data packet in the data stream is encoded separately to obtain a corresponding code block or coded data packet. The coding increases the reliability of the data transmission. The coding scheme may include cyclic redundancy check (CRC) generation, convolutional coding, Turbo coding, low density parity check (LDPC) coding, block coding, other coding, or a combination thereof. With spatial spreading, the SNR can vary across a code block even if the MIMO channel is static over the code block. A sufficiently powerful coding scheme may be used to combat the SNR variation across the code block, so that coded performance is proportional to the average SNR across the code block. Some exemplary coding schemes that can provide good performance for spatial spreading include Turbo code (e.g., the one defined by IS-856), LDPC code, and convolutional code.
• A channel interleaver 624 interleaves (i.e., reorders) the code bits based on an interleaving scheme to achieve frequency, time and/or spatial diversity. The interleaving may be performed across a code block, a partial code block, multiple code blocks, and so on. A symbol mapping unit 626 maps the interleaved bits based on a modulation scheme and provides a stream of data symbols {s _ℓ}. Unit 626 groups each set of B interleaved bits to form a B-bit value, where B ≥ 1, and further maps each B-bit value to a specific modulation symbol based on the modulation scheme (e.g., QPSK, M-PSK, or M-QAM, where M = 2^B). Unit 626 provides a block of data symbols for each code block.
• In FIG. 6, N_D TX data stream processors 620 process N_D data streams. One TX data stream processor 620 may also process the N_D data streams, e.g., in a time division multiplex (TDM) manner.
• Data may be transmitted in various manners in the MIMO system. For example, if N_D = 1, then one data stream is processed, demultiplexed, and transmitted on all N_S transmission channels of the MIMO channel. If N_D = N_S, then one data stream may be processed and transmitted on each transmission channel. In any case, the data to be sent on each transmission channel may be encoded and modulated based on the rate selected for that transmission channel. A multiplexer/demultiplexer (Mux/Demux) 628 receives and multiplexes/demultiplexes the data symbols for the N_D data streams into N_S data symbol streams, one data symbol stream for each transmission channel. If N_D = 1, then Mux/Demux 628 demultiplexes the data symbols for one data stream into N_S data symbol streams. If N_D = N_S, then Mux/Demux 628 can simply provide the data symbols for each data stream as a respective data symbol stream.
• TX spatial processor 530 receives and spatially processes the N_S data symbol streams. Within TX spatial processor 530, a spatial spreader 632 receives the N_S data symbol streams, performs spatial spreading for each transmission span m with the steering matrix V (m) selected for that transmission span, and provides N_S spread symbol streams. The steering matrices may be retrieved from a steering matrix (SM) storage 642 within memory unit 542 or generated by controller 540 as they are needed. A spatial processor 634 then spatially processes the N_S spread symbol streams with the identity matrix I for partial-CSI transmission or with the matrices E (m) of eigenvectors for full-CSI transmission. A multiplexer 636 multiplexes the transmit symbols from spatial processor 634 with pilot symbols (e.g., in a time division multiplexed manner) and provides N_T transmit symbol streams for the N_T transmit antennas.
• FIG. 7 shows a block diagram of an RX spatial processor 560a and an RX data processor 570a, which are one embodiment of RX spatial processor 560 and RX data processor 570, respectively, at receiving entity 150. N_R receiver units 554a through 554r provide received pilot symbols, $\left\{r_i^p\right\}$

  

  for i = 1 ... N_R, to channel estimator 584. Channel estimator 584 estimates the channel response matrix H (m) based on the received pilot symbols and further estimates the SNR of each transmission channel. Controller 580 derives a spatial filter matrix M (m) and possibly a diagonal matrix D (m) for each transmission span m based on the channel response matrix H (m) and possibly the steering matrix V (m). Receiving entity 150 is synchronized with transmitting entity 110 so that both entities use the same steering matrix V (m) for each transmission span m. The matrix M (m) may be derived as shown in equation (10) for the full-CSI transmission and as shown in equations (17) and (23) for the partial-CSI transmission with the CCMI and MMSE techniques, respectively. The matrix M (m) may or may not include the steering matrix V (m) depending on whether the receiver spatial processing and spatial despreading are performed jointly or separately.
• FIG. 7 shows receiver spatial spreading and spatial despreading being performed separately. RX spatial processor 560 obtains received data symbols, $\left\{r_i^d\right\}$

  

  for i=1 ... N_R, from receiver units 554a through 554r and the matrices M (m) and V (m) from controller 580. Within RX spatial processor 560, a spatial processor 762 performs receiver spatial processing on the received data symbols for each transmission span with the matrices M (m). A spatial despreader 764 then performs spatial despreading with the matrix V (m) and provides recovered symbols to RX data processor 570. The receiver spatial processing and spatial despreading may also be performed jointly using the effective MIMO channel estimate, as described above.
• For the embodiment shown in FIG. 7, RX data processor 570a includes a multiplexer/demultiplexer (Mux/Demux) 768 and N_D RX data stream processors 770a through 770nd for the N_D data streams. Mux/Demux 768 receives and multiplexes/demultiplexes the N_S recovered symbol streams for the N_S transmission channels into N_D recovered symbol streams for the N_D data streams. Within each RX data stream processor 770, a symbol demapping unit 772 demodulates the recovered symbols for its data stream in accordance with the modulation scheme used for that stream and provides demodulated data. A channel deinterleaver 774 deinterleaves the demodulated data in a manner complementary to the interleaving performed on that stream by transmitting entity 110. A decoder 776 decodes the deinterleaved data in a manner complementary to the encoding performed by transmitting entity 110 on that stream. For example, a Turbo decoder or a Viterbi decoder may be used for decoder 776 if Turbo or convolutional coding, respectively, is performed by transmitting entity 110. Decoder 776 provides a decoded data stream, which includes a decoded data packet for each data symbol block.
• FIG. 8 shows a block diagram of an RX spatial processor 560b and an RX data processor 570b, which implement the SIC technique for receiving entity 150. For simplicity, N_D = N_S and RX spatial processor 560b and RX data processor 570b implement N_S cascaded receiver processing stages for the N_S data symbol streams. Each of stages 1 to N_S-1 includes a spatial processor 860, an interference canceller 862, an RX data stream processor 870, and a TX data stream processor 880. The last stage includes only a spatial processor 860ns and an RX data stream processor 870ns. Each RX data stream processor 870 includes a symbol demapping unit, a channel deinterleaver, and a decoder, as shown in FIG. 7. Each TX data stream processor 880 includes an encoder, a channel interleaver, and a symbol mapping unit, as shown in FIG. 6.
• For stage 1, spatial processor 860a performs receiver spatial processing on the N_R received symbol streams and provides one recovered symbol stream {ŝ ₁}. RX data stream processor 870a demodulates, deinterleaves, and decodes the recovered symbol stream {ŝ ₁} and provides a corresponding decoded data stream {d̂ ₁}. TX data stream processor 880a encodes, interleaves, and modulates the decoded data stream {d̂ ₁} in the same manner performed by transmitting entity 110 for that stream and provides a remodulated symbol stream {

  

  }. Interference canceller 862a spatially spreads the remodulated symbol stream {

  

  } with the steering matrix V (m) and further multiplies the results with the channel response matrix Ĥ (m) to obtain N_R interference components due to data symbol stream {s ₁}. The N_R interference components are subtracted from the N_R received symbol streams to obtain N_R modified symbol streams, which are provided to stage 2.
• Each of stages 2 through N_S-1 performs the same processing as stage 1, albeit on the N_R modified symbol streams from the preceding stage instead of the N_R received symbol streams. The last stage performs spatial processing and decoding on the N_R modified symbol streams from stage N_S-1 and does not perform interference estimation and cancellation.
• Spatial processors 860a through 860ns may each implement the CCMI, MMSE, or some other technique. Each spatial processor 860 multiplies an input (received or modified) symbol vector ${\underline{\mathbf{r}}}_{\mathit{sic}}^{\ell }\left(m\right)$

  

  with a spatial filter matrix ${\underline{\mathbf{M}}}_{\mathit{sic}}^{\ell }\left(m\right)$

  

  and the steering matrix V (m) to obtain a recovered symbol vector ${\underline{\hat{\mathbf{s}}}}_{\mathit{sic}}^{\ell }\left(m\right)$

  

  and provides the recovered symbol stream for that stage. The matrix ${\underline{\mathbf{M}}}_{\mathit{sic}}^{\ell }\left(m\right)$

  

  is derived based on a reduced channel response matrix Ĥ ^ℓ(m) for the stage. The matrix Ĥ ^ℓ(m) is equal to Ĥ (m) with the columns for all of the data symbol streams already recovered in prior stages removed.
4. Rate Selection and Control
• For both full-CSI and partial-CSI transmission, the receiving entity can estimate the SNR of each transmission channel. The SNR computation is dependent on (1) whether full-CSI or partial-CSI transmission is used, (2) whether spatial spreading is performed, and (3) the particular receiver spatial processing technique (e.g., CCMI, MMSE, or SIC) used by the receiving entity in the case of partial-CSI transmission. For a MIMO-OFDM system, the SNR for each subband of each transmission channel may be estimated and averaged to obtain the SNR of the transmission channel. In any case, an operating SNR, γ_op (ℓ), for each transmission channel may be computed based on the SNR of the transmission channel, γ_pd (ℓ), and an SNR offset, γ_os (ℓ), as follows: $${\gamma }_{\mathit{op}}\left(\ell \right)={\gamma }_{\mathit{pd}}\left(\ell \right)+{\gamma }_{\mathit{os}}\left(\ell \right),$$

  

  where the units are in decibels (dB). The SNR offset may be used to account for estimation error, variability in the channel, and other factors. A suitable rate is selected for each transmission channel based on the operating SNR of the transmission channel.
• The MIMO system may support a specific set of rates. One of the supported rates may be for a null rate, which is a data rate of zero. Each of the remaining rates is associated with a particular non-zero data rate, a particular coding scheme or code rate, a particular modulation scheme, and a particular minimum SNR required to achieve a desired level of performance, e.g., 1% packet error rate (PER) for a non-fading AWGN channel. For each supported non-zero rate, the required SNR may be obtained based on the specific system design (such as the particular code rate, interleaving scheme, and modulation scheme used by the system for that rate) and for an AWGN channel. The required SNR may be obtained by computer simulation, empirical measurements, and so on, as is known in the art. The set of supported rates and their required SNRs may be stored in a look-up table.
• The operating SNR, γ_op (ℓ), of each transmission channel may be provided to the look-up table, which then returns the rate q(ℓ) for that transmission channel. This rate is the highest supported rate with a required SNR, γ_req (ℓ), that is less than or equal to the operating SNR, or γ_req (ℓ)≤γ_op (ℓ). The receiving entity can thus select the highest possible rate for each transmission channel based on its operating SNR.
5. Steering Matrix Generation
• The steering matrices used for spatial spreading may be generated in various manners, and some exemplary schemes are described below. A set of L steering matrices may be pre-computed and stored at the transmitting and receiving entities and thereafter retrieved for use as they are needed. Alternatively, these steering matrices may be computed in real time as they are needed.
• The steering matrices should be unitary matrices and satisfy the following condition: $${\underline{\mathbf{V}}}^H\left(i\right)\cdot \underline{\mathbf{V}}\left(i\right)=\underline{\mathbf{I}},\mathrm{for}i=1\dots L\mathrm{.}$$

  
• Equation (28) indicates that each column of V (i) should have unit energy and the Hermitian inner product of any two columns of V (i) should be zero. This condition ensures that the N_S data symbols sent simultaneously using the steering matrix V (i) have the same power and are orthogonal to one another prior to transmission.
• Some of the steering matrices may also be uncorrelated so that the correlation between any two uncorrelated steering matrices is zero or a low value. This condition may be expressed as: $$\underline{\mathbf{C}}\left(\mathit{ij}\right)={\underline{\mathbf{V}}}^H\left(i\right)\cdot \underline{\mathbf{V}}\left(i\right)\approx \underline{0},\mathrm{for}i=1\dots \mathrm{L},\mathit{j}=1\dots L,\mathrm{and}i\ne j,$$

  

  where C (ij) is the correlation matrix for V (i) and V (j) and 0 is a matrix of all zeros. The condition in equation (30) may improve performance for some applications but is not necessary for most applications.
• The set of L steering matrices { V} may be generated using various schemes. In a first scheme, the L steering matrices are generated based on matrices of random variables. An N_S × N_T matrix G with elements that are independent identically distributed complex Gaussian random variables, each having zero mean and unit variance, is initially generated. An N_T × N_T correlation matrix of G is computed and decomposed using eigenvalue decomposition as follows: $${\underline{\mathbf{R}}}_{\mathrm{G}}={\underline{\mathbf{G}}}^H\cdot \underline{\mathbf{G}}={\underline{\mathbf{E}}}_{\mathrm{G}}\cdot {\underline{\mathbf{D}}}_{\mathrm{G}}\cdot {\underline{\mathbf{E}}}_{\mathrm{G}}^H\mathrm{.}$$

  
• The matrix E _G is used as a steering matrix V (i) and added to the set. The process is repeated until all L steering matrices are generated.
• In a second scheme, the L steering matrices are generated based on a set of (log₂ L)+1 independent isotropically distributed (IID) unitary matrices, as follows: $$\underline{\mathbf{V}}\left({\ell }_1{<figure-callout\; id="2"\; label="l"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\ell </figure-callout>}_2\dots {\ell }_{\mathrm{Q}}\right)={\underline{\mathrm{\Omega }}}_1^{{<figure-callout\; id="1"\; label="l"\; filenames="imgf0006.png,imgf0007.png"\; state="\{\{state\}\}">\ell </figure-callout>}_1}\cdot {\underline{\mathrm{\Omega }}}_2^{{<figure-callout\; id="2"\; label="l"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\ell </figure-callout>}_2}\cdot \dots \cdot {\underline{\mathrm{\Omega }}}_{\mathrm{Q}}^{{\ell }_{\mathrm{Q}}}\cdot {\underline{\mathbf{V}}}_0,\mathrm{for}{<figure-callout\; id="1"\; label="l"\; filenames="imgf0006.png,imgf0007.png"\; state="\{\{state\}\}">\ell </figure-callout>}_1,{<figure-callout\; id="2"\; label="l"\; filenames="imgf0008.png"\; state="\{\{state\}\}">\ell </figure-callout>}_2,\dots ,{\ell }_{\mathrm{Q}}\in \left\{0,1\right\},$$

  
• where V ₀ is an N_T × N_S independent isotropically distributed unitary matrix;
• i = ℓ,ℓ₂...ℓ_Q, where Q = log₂ L and ℓ _j , is the j-th bit of index i; and
• ${\underline{\mathbf{\Omega }}}_j^{{\ell }_j},$
  
  for j = 1 ... Q, is an N_T × N_T IID unitary matrix.
• The second scheme is described by T.L. Marzetta et al. in "Structured Unitary Space-Time Autocoding Constellations," IEEE Transaction on Information Theory, Vol. 48, No. 4, April 2002.
• In a third scheme, the L steering matrices are generated by successively rotating an initial unitary steering matrix V (1) in an N_T-dimensional complex space, as follows: $$\underline{\mathbf{V}}\left(i+1\right)={\underline{\mathrm{\Theta }}}^i\cdot \underline{\mathbf{V}}\left(i\right),\mathrm{for}i=1\dots L-1,$$

  

  where Θ ⁱ is an N_T × N_T diagonal unitary matrix with elements that are L-th roots of unity. The third scheme is described by B.M. Hochwald et al. in "Systematic Design of Unitary Space-Time Constellations," IEEE Transaction on Information Theory, Vol. 46, No. 6, September 2000.
• In a fourth scheme, the set of L steering matrices is generated with a base matrix B and different scalars. The base matrix may be a Walsh matrix, a Fourier matrix, or some other matrix. A 2 × 2 Walsh matrix may be expressed as ${\underline{\mathbf{W}}}_{2\times 2}=\left[\begin{array}{cc}1& \hfill 1\\ 1& \hfill -1\end{array}\right]\mathrm{.}$

  

  A larger size Walsh matrix W _2N×2N may be formed from a smaller size Walsh matrix W _N×N, as follows: $${\underline{\mathbf{W}}}_{2\mathrm{N}\times \mathrm{2}\mathrm{N}}=\left[\begin{array}{cc}{\underline{\mathbf{W}}}_{\mathrm{N}\times \mathrm{N}}& {\underline{\mathbf{W}}}_{\mathrm{N}\times \mathrm{N}}\\ {\underline{\mathbf{W}}}_{\mathrm{N}\times \mathrm{N}}& -{\underline{\mathbf{W}}}_{\mathrm{N}\times \mathrm{N}}\end{array}\right]\mathrm{.}$$

  
• Walsh matrices have dimensions that are powers of two.
• An N_T × N_T Fourier matrix D has element w_n,m in the n-th row of the m-th column, which may be expressed as: $$w_{n,m}=e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="imgf0008.png"\; state="\{\{state\}\}">j</figure-callout>}2\pi \frac{\left(n-1\right)\left(m-1\right)}{{\mathrm{N}}_{\mathrm{T}}}},\mathrm{for}n=\left\{1\dots {\mathrm{N}}_{\mathrm{T}}\right\}\mathrm{and}m=\left\{1\dots {\mathrm{N}}_{\mathrm{T}}\right\},$$

  

  where n is a row index and m is a column index. Fourier matrices of any square dimension (e.g., 2, 3, 4, 5, and so on) may be formed.
• An N_T × N_T Walsh matrix W , Fourier matrix D , or some other matrix may be used as the base matrix B to form other steering matrices. Each of rows 2 through N_T of the base matrix may be independently multiplied with one of M different possible scalars, where M > 1. M^NT-1 different steering matrices may be obtained from M^NT-1 different permutations of the M scalars for the N_T-1 rows. For example, each of rows 2 through N_T may be independently multiplied with a scalar of +1, -1, +j, or -j, where $j=\sqrt{-1}\mathrm{.}$

  

  For N_T = 4 and M = 4, 64 different steering matrices may be generated from the base matrix B with the four different scalars. Additional steering matrices may be generated with other scalars, e.g., e ^±j3π/4, e ^±jπ/4, e ^±jπ/8, and so on. In general, each row of the base matrix may be multiplied with any scalar having the form e^jθ, where θ may be any phase value. N_T × N_T steering matrices may be generated as V (i) = g _NT · B (i), where ${\mathit{g}}_{{\mathrm{N}}_{\mathrm{T}}}=\mathrm{1}/\sqrt{{\mathrm{N}}_{\mathrm{T}}}$

  

  and B(i) is the i-th matrix generated with the base matrix B . The scaling by g _NT ensures that each column of V (i) has unit power.
• Other schemes may also be used to generate the set of L steering matrices, and this is within the scope of the invention. In general, the steering matrices may be generated in a pseudo-random manner (e.g., such as the first scheme) or a deterministic manner (e.g., such as the second, third, and fourth schemes).
• The spatial spreading techniques described herein may be implemented by various means. For example, these techniques may be implemented in hardware, software, or a combination thereof. For a hardware implementation, the processing units for spatial spreading at the transmitting entity and spatial despreading at the receiving entity may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof.
• For a software implementation, the spatial spreading techniques may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes may be stored in memory units (e.g., memory units 542 and 582 in FIG. 5) and executed by a processor (e.g., controllers 540 and 580 in FIG. 5). The memory unit may be implemented within the processor or external to the processor, in which case it can be communicatively coupled to the processor via various means as is known in the art.
• Headings are included herein for reference and to aid in locating certain sections. These headings are not intended to limit the scope of the concepts described therein under, and these concepts may have applicability in other sections throughout the entire specification.
• The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (51)

1. A method (300) of transmitting data from a transmitting entity to a receiving entity in a wireless multiple-input multiple-output, MIMO, communication system, comprising:

   processing (312, 314) data to obtain a plurality of streams of data symbols for transmission on a plurality of transmission channels in a MIMO channel between the transmitting entity and the receiving entity;

   performing (316) spatial spreading on the plurality of streams of data symbols with a plurality of steering matrices to obtain a plurality of streams of spread symbols, wherein the spatial spreading comprises transmission of the data symbol on multiple eigenmodes or spatial channels of the MIMO channel simultaneously with the steering vector;

   performing (318) spatial processing on the plurality of streams of spread symbols for either full channel state information, CSI, transmission on multiple eigenmodes of the MIMO channel or partial-CSI transmission on spatial channels of the MIMO channel to obtain a plurality of streams of transmit symbols for transmission from a plurality of transmit antennas at the transmitting entity.
2. The method of claim 1, wherein the performing spatial processing comprises
   multiplying the plurality of streams of spread symbols with matrices of eigenvectors to transmit the plurality of streams of spread symbols on a plurality of eigenmodes of the MIMO channel.
3. The method of claim 1, wherein the performing spatial processing comprises
   providing each of the plurality of streams of spread symbols as one of the plurality of streams of transmit symbols.
4. The method of claim 1, wherein the processing data comprises
   encoding and modulating data for each of the plurality of streams of data symbols based on a rate selected for the stream of data symbols.
5. The method of claim 4, further comprising:

   obtaining the rate for each stream of data symbols, the rate being selected based on a signal-to-noise-and-interference ratio ,SNR, of a transmission channel for the stream of data symbols.
6. The method of claim 1, wherein the processing data comprises
   encoding and modulating each of a plurality of packets of data to obtain a block of data symbols, and
   multiplexing a plurality of blocks of data symbols generated for the plurality of packets of data onto the plurality of streams of data symbols.
7. The method of claim 6, wherein the encoding and modulating comprise
   encoding each packet of data based on a Turbo code, a convolutional code, or a low density parity check, LDPC, code to obtain a block of coded data, and
   symbol mapping each block of coded data based on a modulation scheme to obtain a block of data symbols.
8. The method of claim 6, wherein the multiplexing the plurality of blocks of data symbols comprises
   multiplexing each block of data symbols onto one of the plurality of streams of data symbols.
9. The method of claim 6, wherein the multiplexing the plurality of blocks of data symbols comprises
   multiplexing each block of data symbols onto all of the plurality of streams of data symbols.
10. The method of claim 6, wherein the performing spatial spreading comprises
    performing spatial processing on each block of data symbols in the plurality of streams of data symbols with at least two steering matrices.
11. The method of claim 1, wherein the performing spatial spreading comprises
    performing spatial processing on the plurality of streams of data symbols using a set of L steering matrices, where L is an integer greater than one.
12. The method of claim 11, further comprising:

    generating the L steering matrices as unitary matrices having orthogonal columns.
13. The method of claim 11, further comprising:

    selecting a steering matrix from among the L steering matrices for each time interval, and wherein the spatial spreading is performed for each time interval with the steering matrix selected for the time interval.
14. The method of claim 11, further comprising:

    selecting a steering matrix from among the L steering matrices for each group of at least one frequency subband, and wherein the spatial spreading is performed for each group of at least one frequency subband with the steering matrix selected for the group.
15. The method of claim 1, further comprising:

    processing each of the plurality of streams of transmit symbols for orthogonal frequency division multiplexing, OFDM,.
16. The method of claim 1, wherein the performing spatial spreading comprises
    performing spatial processing with at least two different steering matrices for a plurality of subbands in each symbol period with data transmission.
17. An apparatus (110) in a wireless multiple-input multiple-output ,MIMO, communication system, comprising:

    means for processing (312, 314) data to obtain a plurality of streams of data symbols for transmission on a plurality of transmission channels in a MIMO channel between a transmitting entity and a receiving entity in the MIMO system;

    means for performing (316) spatial spreading on the plurality of streams of data symbols with a plurality of steering matrices to obtain a plurality of streams of spread symbols, wherein the spatial spreading comprises transmission of the data symbol on multiple eigenmodes or spatial channels of a MIMO channel simultaneously with the steering vector;

    means for performing (318) spatial processing on the plurality of streams of spread symbols for either full channel state information, CSI, transmission on multiple eigenmodes of the MIMO channel or partial-CSI transmission on spatial channels of the MIMO channel to obtain a plurality of streams of transmit symbols for transmission from a plurality of transmit antennas at the apparatus.
18. The apparatus of claim 17, wherein the means for performing spatial processing comprises
    means for multiplying the plurality of streams of spread symbols with matrices of eigenvectors to transmit the plurality of streams of spread symbols on a plurality of eigenmodes of the MIMO channel.
19. The apparatus of claim 17, wherein the means for performing spatial processing comprises
    means for providing each of the plurality of streams of spread symbols as one of the plurality of streams of transmit symbols.
20. The apparatus of claim 17, wherein the means for processing data comprises
    means for encoding and modulating data for each of the plurality of streams of data symbols in accordance with a rate selected based on a signal-to-noise-and interference ratio, SNR, of a transmission channel for the stream of data symbols.
21. The apparatus of claim 17, wherein the means for processing data comprises
    means for encoding and modulating each of a plurality of packets of data to obtain a block of data symbols, and
    means for multiplexing a plurality of blocks of data symbols generated for the plurality of packets of data onto the plurality of streams of data symbols.
22. The apparatus of claim 21, wherein the means for performing spatial spreading spatially spreads each block of data symbols in the plurality of streams of data symbols with at least two steering matrices.
23. The apparatus of claim 17, further comprising:

    means for selecting a steering matrix from among L steering matrices for each time interval, where L is an integer greater than one, and wherein the spatial spreading for each time interval is performed with the steering matrix selected for the time interval.
24. The apparatus of claim 17, wherein the MIMO system utilizes orthogonal frequency division multiplexing, OFDM, and wherein means for performing spatial spreading comprises
    means for performing spatial spreading with at least two different steering matrices for a plurality of subbands in each symbol period with data transmission.
25. The apparatus of claim 17, wherein the spatial spreading performed by the means for performing spatial spreading results in whitened interference and noise observed by the receiving entity for the plurality of streams of data symbols after spatial despreading by the receiving entity.
26. The apparatus of claim 17, wherein the MIMO channel includes plurality of spatial channels, and wherein the spatial spreading performed by the means for performing spatial spreading results in each of the plurality of transmission channels achieving a signal-to-noise-and interference ratio ,SNR, that is an average of SNRs of the plurality of spatial channels.
27. The apparatus according to any of claims 17 to 26, wherein
    the means for processing data comprises a data processor;
    the means for performing spatial spreading comprises a spatial spreader, and
    the means for performing spatial processing comprises a spatial processor.
28. A method (400) of receiving a data transmission sent by a transmitting entity to a receiving entity in a wireless multiple-input multiple-output, MIMO, communication system, comprising:

    obtaining (412) a plurality of streams of received symbols for a plurality of streams of data symbols transmitted via a plurality of transmission channels in a MIMO channel, wherein the plurality of streams of data symbols are spatially spread with a plurality of steering matrices and further spatially processed prior to transmission via the MIMO channel, and wherein the spatial spreading comprises transmission of the data symbol on multiple eigenmodes or spatial channels of a MIMO channel simultaneously with the steering vector; estimating (414) the MIMO channel response; performing (416) full channel state information, CSI, or partial-CSI receiver spatial processing on the plurality of streams of received symbols with the MIMO channel estimate to obtain a plurality of streams of detected symbols; and performing (418) spatial despreading on the plurality of streams of detected symbols with the plurality of steering matrices to obtain a plurality of streams of recovered symbols, which are estimates of the plurality of streams of data symbols.
29. The method of claim 28, further comprising:

    obtaining an effective MIMO channel estimate that includes a channel response estimate for the MIMO channel and the plurality of steering matrices used for spatial spreading; and

    performing the receiver spatial processing and the spatial despreading jointly based on the effective MIMO channel estimate.
30. The method of claim 28, wherein the performing spatial despreading comprises
    performing spatial despreading on each block of detected symbols in the plurality of streams of detected symbols with at least two steering matrices used by the transmitting entity on a corresponding block of data symbols.
31. The method of claim 28, wherein the performing spatial despreading comprises
    performing spatial despreading on the plurality of streams of detected symbols using a set of L steering matrices, where L is an integer greater than one, and wherein the L steering matrices are unitary matrices.
32. The method of claim 28, wherein the performing receiver spatial processing comprises
    multiplying the plurality of streams of received symbols with matrices of eigenvectors for a plurality of eigenmodes of the MIMO channel to obtain the plurality of streams of detected symbols.
33. The method of claim 28, wherein the performing receiver spatial processing comprises
    deriving a matched filter based on a channel response estimate for the MIMO channel, and
    multiplying the plurality of streams of received symbols with the matched filter to obtain the plurality of streams of detected symbols.
34. The method of claim 28, wherein the performing receiver spatial processing comprises
    performing receiver spatial processing on the plurality of streams of received symbols based on a channel correlation matrix inversion, CCMI, technique.
35. The method of claim 28, wherein the performing receiver spatial processing comprises
    performing receiver spatial processing on the plurality of streams of received symbols based on a minimum mean square error, MMSE, technique.
36. The method of claim 28, wherein the performing receiver spatial processing comprises
    performing receiver spatial processing on the plurality of streams of received symbols based on a successive interference cancellation ,SIC, technique.
37. The method of claim 28, further comprising:

    estimating a signal-to-noise-and-interference ratio ,SNR, of each of the plurality of transmission channels for the plurality of streams of data symbols; and

    selecting a rate for each of the plurality of streams of data symbols based on an SNR estimate for the transmission channel for the stream of data symbols.
38. The method of claim 28, further comprising:

    sending to the transmitting entity at least one rate for the plurality of streams of data symbols, wherein the plurality of streams of data symbols are encoded and modulated based on the at least one rate.
39. The method of claim 28, further comprising:

    estimating a signal-to-noise-and-interference ratio, SNR, of each of the plurality of transmission channels for the plurality of streams of data symbols; and

    selecting a single rate for the plurality of streams of data symbols based on SNR estimates for the plurality of transmission channels.
40. The method of claim 28, further comprising:

    demodulating and decoding each of the plurality of streams of recovered symbols based on a rate selected for the stream to obtain decoded data.
41. The method of claim 28, wherein the performing spatial despreading comprises
    performing spatial despreading on the plurality of streams of detected symbols with at least two steering matrices for a plurality of subbands of each symbol period used for data transmission.
42. An apparatus (150) in a wireless multiple-input multiple-output, MIMO, communication system, comprising:

    means for obtaining (412) a plurality of streams of received symbols for a plurality of streams of data symbols transmitted via a plurality of transmission channels in a MIMO channel from a transmitting entity to a receiving entity, wherein the plurality of streams of data symbols are spatially spread with a plurality of steering matrices and further spatially processed prior to transmission via the MIMO channel, and wherein the spatial spreading comprises transmission of the data symbol on multiple eigenmodes or spatial channels of a MIMO channel simultaneously with the steering vector; means for estimating (414) the MIMO channel response; means for performing (416) full channel state information, CSI, or partial-CSI receiver spatial processing on the plurality of streams of received symbols with the MIMO channel estimate to obtain a plurality of streams of detected symbols; and

    means for performing (418) spatial despreading on the plurality of streams of detected symbols with the plurality of steering matrices to obtain a plurality of streams of recovered symbols, which are estimates of the plurality of streams of data symbols.
43. The apparatus of claim 42, further comprising:

    a channel estimator to obtain an effective MIMO channel estimate that includes a channel response estimate for the MIMO channel and the plurality of steering matrices used for spatial spreading, and wherein the means for performing receiver spatial processing and the means for performing spatial despreading perform receiver spatial processing and spatial despreading jointly based on the effective MIMO channel estimate.
44. The apparatus of claim 42, wherein the means for performing receiver spatial processing comprises means for multiplying the plurality of streams of received symbols with matrices of eigenvectors for a plurality of eigenmodes of the MIMO channel to obtain the plurality of streams of detected symbols.
45. The apparatus of claim 42, wherein the means for performing receiver spatial processing comprises means for multiplying the plurality of streams of received symbols with a matched filter, derived based on a channel response estimate for the MIMO channel, to obtain the plurality of streams of detected symbols.
46. The apparatus of claim 42, wherein the means for performing receiver spatial processing performs receiver spatial processing based on a channel correlation matrix inversion, CCMI, technique, a minimum mean square error, MMSE technique, or a successive interference cancellation, SIC, technique.
47. The apparatus of claim 42, further comprising: means for estimating a signal-to-noise-and-interference ratio, SNR of each of the plurality of transmission channels for the plurality of streams of data symbols; and
    means for selecting a rate for each of the plurality of streams of data symbols based on an SNR estimate for a transmission channel for the stream of data symbols, and wherein each stream of data symbols is encoded and modulated by the transmitting entity based on the rate selected for the stream of data symbols.
48. The apparatus of claim 47, wherein the means for estimating the SNR comprises a channel estimator; and
    the means for selecting comprises a controller.
49. The apparatus of claim 42, further comprising:

    a data processor to demodulate and decode each of the plurality of streams of recovered symbols based on a rate selected for the stream to obtain decoded data.
50. The apparatus of claim 42, wherein the MIMO system utilizes orthogonal frequency division multiplexing, OFDM, and wherein the spatial despreader performs spatial despreading with at least two different steering matrices for a plurality of subbands in each symbol period with data transmission.
51. A computer readable medium comprising instructions that when executed by a processor perform the steps of any one of claims 1 to 16 or 28 to 41.

Images